Office and 24-h ambulatory blood pressure control by treatment in general practice: The 'Monitoraggio della pressione ARteriosa nella medicina TErritoriale' study

Background: Guidelines recommend that blood pressure (BP) should be lowered in hypertensive patients to prevent cardiovascular accidents. Management of antihypertensive treatment by general practitioners is usually based on office measurements, which may not allow an assessment of BP control over 24 h, which requires ambulatory BP monitoring (ABPM) to be implemented. This is rarely done in general practice, and limited information is available on the consistency between the evaluations of the response to treatment provided by office measurement and by ABPM in this setting. AIM: To assess concordance between office BP measurements and ABPM-based estimates of hypertension control in a general practice setting. Design of Study: Prospective, comparative between techniques. Setting: General practice. Methods: Seventy-eight general practices, representative of all Italian regions, participated in this study by recruiting sequential hypertensive adults on stabilized treatment, who were subdivided into even groups with office BP, respectively, controlled or noncontrolled by treatment. In each individual, ABPM was applied by the general practitioner after appropriate training, and 24-h ABP values were defined as controlled or not according to current guidelines. Concordance between office and ABPM evaluation of BP control was assessed with κ statistics. Positive and negative predictive values of office measurement versus ABPM were estimated. Results: Between July 2005 and November 2006, 190 general practitioners recruited 2059 hypertensive patients based on office BP measurements; in 1728 patients, a 24-h ABPM was performed, yielding 1524 recordings considered as valid for further analysis. The agreement between the assessment of BP control by office measurement and by ABPM was poor (κ = 0.120), with office measurements showing a satisfactory positive predictive value (0.842) and a poor negative predictive value (0.278); the situation was worse in patients with three or more among the following features: male sex, age of at least 65 years, alcohol consumption, diabetes, and obesity (negative predictive value = 0.149). Conclusion: In general practice, the agreement between assessment of BP control by treatment provided by office and ambulatory BP measurements is better in patients of 'uncontrolled' office BP than in 'controlled' office BP patients. This emphasizes the need for the larger use of out-of-office BP monitoring in a general practice setting, in particular, in patients considered as 'controlled' during consultation.